Was there any test/check on the console verification movie vs. the mupen run (using whatever exact framerate mupen is trying to emulate), or is it possible to skip those approximation and maybe the n64 bot can calculate the amount of lag frame the n64 doesn't poll controller updates?
While viewing the 0Star run on SonicPacker's channel and the console verification side-by-side, it looks like the console version gets ahead. Overall the difference is small, but on console seems slightly faster (about 1/2 second). Once I find a way for me to record the N64 video, I'll create a side-by-side, frame-by-frame video of the console recording and the emulator recording for 0Star. The N64 framerate is very jumpy, and dips below 30fps quite often, so idk if a recording would skew the timing on video playback, would it?
I've heard of speedrunners recording their runs to DVD. You could try that (assuming you have a DVD player), and then you wouldn't have to buy any additional recording hardware/software.
SoulCal wrote:
The N64 framerate is very jumpy, and dips below 30fps quite often, so idk if a recording would skew the timing on video playback, would it?
I wouldn't think so. Real-time speedruns are recorded from the N64 as well (and timed based on that). And I would think that the N64's lag patterns would be constant (deterministic) across multiple replays of the same run. So whatever video you record from your N64 should be of the correct console time.
The N64 framerate is very jumpy, and dips below 30fps quite often, so idk if a recording would skew the timing on video playback, would it?
I wouldn't think so. Real-time speedruns are recorded from the N64 as well (and timed based on that). And I would think that the N64's lag patterns would be constant (deterministic) across multiple replays of the same run. So whatever video you record from your N64 should be of the correct console time.
Do dvd recorders & pvrs add/delete/duplicate frames to lock it to a set framerate? I don't see how else they would create 30fps video from a N64 if the N64 keeps changing to 15-20 fps during lag. If they do, then on playback it'll be the correct time (which is no issue for RTS runs), but not if I want to edit the video to do a frame-by-frame video comparing the console to the emulator. I could be wrong. Idk much about editing.
Do dvd recorders & pvrs add/delete/duplicate frames to lock it to a set framerate? I don't see how else they would create 30fps video from a N64 if the N64 keeps changing to 15-20 fps during lag. If they do, then on playback it'll be the correct time (which is no issue for RTS runs), but not if I want to edit the video to do a frame-by-frame video comparing the console to the emulator. I could be wrong. Idk much about editing.
DVD recorder/pvrs don't care about source fps (nor do they have any idea what the supposed source fps is). They record at their own fps, set by standards or something. Probably something like 24 fps. They don't add/delete/duplicate, they just take whatever the source is feeding them at that exact point in time and encode it, not caring if that picture is the exact same as the last 3 or completely different.
I'm not familliar with console to pc recording box, so I don't know if those alter the input fps before relaying it to the pc, but no matter what, the recording will be at a constant framerate, it doesn't care (or even knows) the current fps the n64 spits out.
DVD recorder/pvrs don't care about source fps (nor do they have any idea what the supposed source fps is). They record at their own fps, set by standards or something. Probably something like 24 fps. They don't add/delete/duplicate, they just take whatever the source is feeding them at that exact point in time and encode it, not caring if that picture is the exact same as the last 3 or completely different.
Not very likely.
We have PAL (50 fps) and NTSC (60 fps) which older consoles output, so that's probably what they record with unless they want to add/drop frames.
For HD, it's a little more complicated, but it's likely to be 30 or 60 fps since those are the most common.
DVD recorder/pvrs don't care about source fps (nor do they have any idea what the supposed source fps is). They record at their own fps, set by standards or something. Probably something like 24 fps. They don't add/delete/duplicate, they just take whatever the source is feeding them at that exact point in time and encode it, not caring if that picture is the exact same as the last 3 or completely different.
Not very likely.
We have PAL (50 fps) and NTSC (60 fps) which older consoles output, so that's probably what they record with unless they want to add/drop frames.
For HD, it's a little more complicated, but it's likely to be 30 or 60 fps since those are the most common.
I might be overlooking something here, movie studio been recording movie in framerate lower then 30 for ages, just a handfull of movie decided to bump it to something like 48 fps (the hobbit is one). DVD standard have been set for a while (and compatibility force them to keep it)
https://en.wikipedia.org/wiki/DVD-Video
If you record using a DVD Recorder box (we are talking on the fly here), they have to follow the standard so the movie can be played back on other DVD players.
HD only means high resolution, blu-ray movies are encoded at 24 fps.
I might be totally wrong on this, but I wouldn't count on it. Pal and NTSC don't mean anything here, a movie can be encoded at the 29.97 standard dvd movie fps, and the box can still send the signal at 60 fps to the tv (no clue if it has to, I'm no video signal wizard).
I might be overlooking something here, movie studio been recording movie in framerate lower then 30 for ages, just a handfull of movie decided to bump it to something like 48 fps (the hobbit is one). DVD standard have been set for a while (and compatibility force them to keep it)
https://en.wikipedia.org/wiki/DVD-Video
Yes, the MOVIE industry does that. Doesn't necessarily mean consumer electronics do.
If you record using a DVD Recorder box (we are talking on the fly here), they have to follow the standard so the movie can be played back on other DVD players.
The only standard that exists AFAIK is the compression standard they use, ie MPEG2. While I am not an expert on the details in the dvd format, I don't think it's limited to 24 fps.
Besides that, I do believe dvd recorders have different options and stuff that may make them incompatible with other players. In this meaning, I think dvd means that it uses dvd as a medium, not that it creates movies that other players may play back (although they probably often can).
HD only means high resolution, blu-ray movies are encoded at 24 fps.
Yes, the term HD refers to resolution, but the standards used for HD is often 720p30, 1080i60 and recently, 1080p60. TV stations tend to broadcast using this. Movies are still 24 fps, though.
I might be totally wrong on this, but I wouldn't count on it. Pal and NTSC don't mean anything here, a movie can be encoded at the 29.97 standard dvd movie fps, and the box can still send the signal at 60 fps to the tv (no clue if it has to, I'm no video signal wizard).
But it does matter. If you don't use the correct standard, the TV cannot interpret it. For older TVs, there was PAL, NTSC and PAL60. That meant 50, 60 and 60 fps, respectively with a little varying resolution. That is the electrical signal the TV could accept. Today is different, though, with TVs being able to accept varying fps, such as 24, 50 and 60, regardless of resolution, all which are popular today.
Because the signal has to be a certain fps, it means that the recorder must change the fps on playback if it isn't the same as the output signal. That means inserted or dropped frames.
What does mupen do when recording video? Does it add/duplicate frames during lag when dumping the .avi?
It would be nice to see the timing differences between console and mupen, so a few frames added here or there in a console recording make a difference, even if the run syncs. I would suspect some areas are faster on console, and vice versa.
What I plan to do is count the # of frames it takes for Mario to, for example, go from the warp pipe to the castle door on the two recordings, and continue this for each area. But of course...I need a way to record.
ToT recently posted some videos.
The Tip Top Of The Huge Island - 16.10(Previous record: 16.20 by sonicpacker & SilentSlayers)Link to videoRoll into the Cage - 6.50(Previous record: 6.57 by SilentSlayers, NEW STARTING STRAT HYPE)Link to videoNew LLL Volcano Entry StratLink to videoFall onto the Caged Island - 8.97(Previous record: 9.20 by USTA)Link to video
would be nice if he had the previous record linked in the description
some of those starting strats are genious, but the old Fall onto the Caged Island record was absolute garbage, I have no idea how it stood for so long O_o
edit: if 16″97 was the old Hot-Foot-It Into The Volcano record then I’m surprised nobody tested to see if the BLJ entry was faster…
Couldn't tell, but I take it that saves time because you skip the fade in sequence?
It doesn't save time sadly, because the 120 uses MIPS #1 to clip through the HMC door.
However, this will save a few frames when fully optimized:
http://www.twitch.tv/ky_man/c/2580749
(Sorry for bad quality)
